Awesome

AdaDSR (ECCV 2020 AIM workshop)

PyTorch implementation of Deep Adaptive Inference Networks for Single Image Super-Resolution

Results

Preparation

• Prerequisites
  • PyTorch (v1.2)
  • Python 3.x, with OpenCV, Numpy, Pillow, tqdm and matplotlib, and tensorboardX is used for visualization
  • [optional] Make sure that matlab is in your PATH, if you want to calculate the PSNR/SSIM indices and use the argument --matlab True
  • [Sparse Conv] cd masked_conv2d; python setup.py install. Note that currently we provide a version modified from open-mmlab/mmdetection, which supports inference with 3x3 sparse convolution layer. We will provide a more general version in the future.
• Dataset
  • Training
    • DIV2K is used for training, you can download the dataset from ETH_CVL or SNU_CVLab.
    • The data should be organized as DIV2K_ROOT/DIV2K_train_HR/*.png, DIV2K_ROOT/DIV2K_train_LR_bicubic/X[234]/*.png, which is identical to the official format.
  • Testing
    • Set5, Set14, B100, Urban100 and Manga109 are used for test.
    • You can download these datasets (~300MB) from Google Drive or Baidu Yun (43qz), and run Prepare_TestData_HR_LR.m in matlab, you can get two folders named HR and LR. Place these two folders in BENCHMARK_ROOT.
  • Models
    • Download the pre-trained models (~2.2GB) from Google Drive or Baidu Yun (cyps), and put the two folders in the root folder.

Quick Start

We show some exemplar commands here for better introduction, and more useful scripts are given in the scripts folder.

Testing

• AdaEDSR

  python test.py --model adaedsr --name adaedsr_x2 --scale 2 --load_path ./ckpt/adaedsr_x2/AdaEDSR_model.pth --dataset_name set5 set14 b100 urban100 manga109 --depth 32 --chop True --sparse_conv True --matlab True --gpu_ids 0
  

• AdaRCAN

  python test.py --model adarcan --name adarcan_x2 --scale 2 --load_path ./ckpt/adarcan_x2/AdaRCAN_model.pth --dataset_name set5 set14 b100 urban100 manga109 --depth 20 --chop True --sparse_conv True --matlab True --gpu_ids 0
  

Training

• AdaEDSR (Load pre-trained EDSR model for more stable training)

  python train.py --model adaedsr --name adaedsr_x2 --scale 2 --load_path ./pretrained/EDSR_official_32_x${scale}.pth
  

• AdaRCAN (Load pre-trained RCAN model for more stable training)

  python train.py --model adarcan --name adarcan_x2 --scale 2 --load_path ./pretrained/RCAN_BIX2.pth
  

Note

• You should set data root by --dataroot DIV2K_ROOT (train) or --dataroot BENCHMARK_ROOT (test), or you can add your own path in the rootlist of div2k_dataset or benchmark_dataset.
• You can specify which GPU to use by --gpu_ids, e.g., --gpu_ids 0,1, --gpu_ids 3, --gpu_ids -1 (for CPU mode). In the default setting, all GPUs are used.
• You can refer to options for more arguments.

Citation

If you find AdaDSR useful in your research, please consider citing:

@inproceedings{AdaDSR,
    title={Deep Adaptive Inference Networks for Single Image Super-Resolution},
    author={Liu, Ming and Zhang, Zhilu and Hou, Liya and Zuo, Wangmeng and Zhang, Lei},
    booktitle={European Conference on Computer Vision Workshops},
    year={2020}
}

Acknowledgement

This repo is built upon the framework of CycleGAN, and we borrow some code from DPSR, mmdetection, EDSR, RCAN and SAN, thanks for their excellent work!